VOGONS


First post, by danijelm

User metadata
Rank Newbie
Rank
Newbie

I've always wanted to see what the difference was between the Intel processors from my high school days. The test platform is ABIT BH6, 128MB 100mhz cl2 ram, geforce2 MX(175/166Mhz), voodoo2 12MB(90Mhz).
Procesors are: P2-450, P3-450 (katmai), P3-600EB(cu-mine), celeron300A. For celeron 300 I disable L2 cache on mendocino.

feel free to comment on the results

I speak sarcasm as a 2nd language

Reply 1 of 9, by Errius

User metadata
Rank l33t
Rank
l33t

This is interesting, I've been meaning to do something similar, but also including the 450 MHz Pentium II Xeon.

Is this too much voodoo?

Reply 2 of 9, by RogueTrip2012

User metadata
Rank Oldbie
Rank
Oldbie

Great work. I appreciate the way you keep the charts colors the same also as it seems so many alway keep me looking at the legend.

Overclock those processors and test again.

I loved my celeron 366@550 on a abit be6-II back in the day till it died.

> W98SE . P3 1.4S . 512MB . Q.FX3K . SB Live! . 64GB SSD
>WXP/W8.1 . AMD 960T . 8GB . GTX285 . SB X-Fi . 128GB SSD
> Win XI . i7 12700k . 32GB . GTX1070TI . 512GB NVME

Reply 3 of 9, by anakin94

User metadata
Rank Newbie
Rank
Newbie

Great to see the little performance boost with the Coppermine.
Good practical test.

Enhanced for Matrox Mystique

Reply 4 of 9, by dionb

User metadata
Rank l33t++
Rank
l33t++

Very interesting to see the difference between software rendering and MX400 - in software rendering the Mendocino pulls clearly ahead of the half-speed cache CPUs, but with a good external GPU the 512kB half speed beats the 128kB full speed.

Reply 5 of 9, by analog_programmer

User metadata
Rank Oldbie
Rank
Oldbie

Nothing surprising as results here, but Voodoo2 is not adequate to be used for testing next to GeForce2. Throw in some TNT(2).

And as always the cacheles Celeron Covington is a piece of pathetic sh*t even at 450 MHz 🤣

The word Idiot refers to a person with many ideas, especially stupid and harmful ideas.
This world goes south since everything's run by financiers and economists.
This isn't voice chat, yet some people overusing online communications talk and hear voices.

Reply 6 of 9, by st31276a

User metadata
Rank Member
Rank
Member

Remember the Katmais have sse, where the Mendocinos dont. Looks like the nvidia driver has an sse code path.

Reply 7 of 9, by dionb

User metadata
Rank l33t++
Rank
l33t++
st31276a wrote on 2024-11-07, 14:24:

Remember the Katmais have sse, where the Mendocinos dont. Looks like the nvidia driver has an sse code path.

Deschutes also doesn't and the difference between Deschutes and Katmai is minimal (a few fps at most) - enough to show the effect of SSE, but Mendocino lags behind Deschutes as well despite supporting the exact same instructions, so cache size/structure is definitely involved.

Reply 8 of 9, by rasz_pl

User metadata
Rank l33t
Rank
l33t
analog_programmer wrote on 2024-11-06, 05:10:

Voodoo2 is not adequate to be used for testing

3dfx driver is super slim and very optimized compared to Nvidia

dionb wrote on 2024-11-07, 15:07:

so cache size/structure is definitely involved

3dfx shows almost no difference, I blame bloated Nvidia driver 😀

https://github.com/raszpl/FIC-486-GAC-2-Cache-Module for AT&T Globalyst
https://github.com/raszpl/386RC-16 memory board
https://github.com/raszpl/440BX Reference Design adapted to Kicad
https://github.com/raszpl/Zenith_ZBIOS MFM-300 Monitor

Reply 9 of 9, by analog_programmer

User metadata
Rank Oldbie
Rank
Oldbie
rasz_pl wrote on 2024-11-08, 02:33:

3dfx driver is super slim and very optimized compared to Nvidia

Voodoo2 card looks like "bottleneck" for some of the system configurations. See that GeForce2 results resemble software rendering picture. I'm just curious if TNT or TNT2 results will be"flat" as those with Voodoo2.

As for the CPUs test I only look at software rendering charts and there's nothing unexpected.

The word Idiot refers to a person with many ideas, especially stupid and harmful ideas.
This world goes south since everything's run by financiers and economists.
This isn't voice chat, yet some people overusing online communications talk and hear voices.